Download Perceptive Images of the Concepts Happiness and Harmony: A

Perceptive Images of the Concepts Happiness and Harmony:
A Comparative Analysis of Data from Russian and English Corpora
Elizaveta Bespalova
National Research University Higher School of Economics
This research is based on the methodology of conceptual metaphor theory (Lakoff and Johnson 1980,
1999; Lakoff 1987; Johnson 1987; Kövecses 2005, 2008, 2013; Gibbs 2006, 79-122; etc.). A
particular issue that the author addresses is perceptive images of abstract concepts (Glebkin 2012,
87-100; Glebkin et al. 2014). From this perspective the main task of research is a detailed description
of a system of perceptive images representing a class of abstract concepts or a particular concept. It
is worth noting that such a trajectory is opposite to a traditional one, which focuses on various aspects
of the path from a primary metaphor to abstract domains that this metaphor represents.
The gist of this paper is a comparative analysis of Russian end English concepts. Two pairs of
concepts are analysed: happiness vs. счастье and harmony vs. гармония. The choice of these
particular concepts follows the fact that the class of abstract concepts expressing feelings is
commonly used and is obvious for both languages, which makes it simple to find exact parallels. The
result of corpus analysis is two pairs of perceptive images. Comparison represents some interesting
results, e.g., in English harmony can be measured in terms of quantity (there can be much less
harmony and lack of harmony) while in Russian it cannot. However, in Russian harmony can be
described with human features, while in English there are no such examples. The differences in
perceptive images for similar concepts in two languages can be explained by cross-cultural
differences: concepts of feelings in Russian are more emotional in comparison to rational ones in
English.
References
Gibbs, R. W. Jr. 2006. Embodiment and cognitive science. Cambridge, New York: Cambridge
University Press. 79-122.
Glebkin, V. 2012. Leksicheskaia semantika: kul`turno-istoricheskii` podhod. M.: Centr gumanitarny`kh
initciativ. 87-100.
Glebkin, V., Aliapkina, A. and Bespalova, E. Pertceptivny`e portrety` abstraktny`kh poniatii`
("schast`e", "svoboda", "beda"). 2014. Vestneyk RGGU, Seriia "Filologicheskie nauki.
Iazy`koznanie", 130, 46-63.
Johnson, M. 1987. The body in the mind. Chicago and London: The University of Chicago Press.
Kövecses, Z. 2005. Metaphor in culture: universality and variation. Cambridge, New York: Cambridge
University Press.
Kövecses, Z. 2008. Conceptual metaphor theory: Some criticisms and alternative proposals. Annual
Review of Cognitive Linguistics, 6, 168–184.
Kövecses, Z. 2013. Recent developments in metaphor theory: Are the new views rival ones? In F.
Gonzálvez-García, M. Peña Cervel and L. Pérez Hernández (Eds.) Metaphor and Metonymy
revisited beyond the Contemporary Theory of Metaphor: Recent developments and
applications (pp. 11–25). Amsterdam: John Benjamins.
Lakoff, G. and Johnson, M. 1980. Metaphors we live by. Chicago and London: The University of
Chicago Press.
Lakoff, G. and Johnson M. 1999. Philosophy in the flesh: the embodied mind and its challenge to
western thought. N.Y.: Basic books.
Lakoff, G. 1987. Women, fire and dangerous things. Chicago and London: The University of Chicago
Press.
Grammatical and Morphological Collocations of Gairaigo in Contemporary Japanese
Anna Bordilovskaya
Kobe University, Graduate School of Humanities
Contemporary Japanese incorporates words of different origin: wago – native words, kango –
Chinese origin words, and gairaigo – words borrowed from the languages other than Chinese. This
results in the increase of near synonymic pairs and rows of different, especially among adjectives.
Most recently, Bordilovskaya (2012) has illustrated that Japanese has a tendency for homogeneity of
origin of adjectives and nouns used in collocations, i.e. gairaigo adjectives are significantly more often
found in collocations with gairaigo nouns rather than wago/kango nouns. In addition, the corpus
search has revealed another tendency: gairaigo adjectives were frequently used with gairaigo nouns
not only in grammatical collocations, but also in a compound-like form.
In this case study, by the means of corpus analysis we test 2 patterns of gairaigo adjective +
gairaigo noun combinations, and investigate which type (grammatical or morphological) is more
common. We have selected 12 gairaigo adjectives: howaito ‘white’, burakku ‘black’, reddo ‘red’, ierō
‘yellow’, buraun ‘brown’, pāpuru ‘purple’, hotto ‘hot’, yangu ‘young’, rongu ‘long’, shōto ‘short’.
Balanced Corpus of Contemporary Written Japanese corpus (BCCWJ) was searched for 2
combinational patterns: Type 1: Morphological Collocations (gairaigo adjective + gairaigo noun); Type
2: Grammatical Collocations (gairaigo modifier + -no/-na connectors + gairaigo noun). The result are
in Table 1.
Table 1. The Distribution of Type 1 and Type 2 Combinational Patterns of Gairaigo Adjectives Use
Gairaigo adjective
Type 1 Morphological Collocations
Type 2 Grammatical Collocations
howaito (‘white’)
burakku (‘black’)
122
62
23
14
reddo (‘red’)
ierō (‘yellow’)
36
97
2
16
buraun (‘brown’)
pāpuru (‘purple’)
46
28
26
12
gurē (‘grey’)
pinku (‘pink’)
48
194
139
347
hotto (‘hot’)
155
6
yangu (‘young’)
48
7
rongu (‘long’)
318
16
shōto (‘short’)
130
12
According to Table 1, Type 2 gairaigo grammatical collocations are outnumbered by Type 1
morphological collocations in 9 cases out of 12, for howaito, burakku, reddo, ierō, buraun, pāpuru,
hotto, yangu, rongu and shōto, i.e., some gairaigo adjectives are more often used in a compound-like
form, than in collocations using -no/-na connectors. These results can give new insights into the
constraints of the use of gairaigo adjectives in Contemporary Japanese, and contribute to the better
understanding of the differentiation between native and borrowed near synonyms.
Highlighting Components of a Frame: A Comparative Study of English and Korean
Youngsoon Cho
Chonnam National University
Cognitive semantics argues that differences in the range and elaboration of metaphor and
those of metonymy show cultural differences in conceptualization (Kövecses 2000). However, studies
on cultural variations in the conceptualization of emotion have focused mainly on differences of
metaphor and have not paid much attention to those of metonymy. This is partly due to the
speculation that metonymy will not demonstrate significant cultural differences because it directly
reflects physiological phenomena, which human beings commonly experience when they feel
emotions, as conceived in Kövecses (2000).
This research aims to show that metonymy also reveals meaningful cultural differences,
through comparing the sadness metonymy of pain in one’s heart, its elaboration, and its use in
English and Korean. In describing the results of this research, the ideas that metonymy is
characterized as highlighting components (Köch 2001) or attention/perspective change towards
components of a frame (Talmy 1996 and Bartsch 2002) will be incorporated.
Observing related data from dictionaries and corpra reveals English and Korean cultures
share the same frame of pain-in-heart with the processes of causing pain, feeling pain, response to
pain, and remaining wounds (Table 1). To compare their uses, a fiction corpus from BNC (16,033,634
words) and one from Sejong (7,096,265 ecel*s) are utilized: From potential data, which was collected
by searching related lemmas occurring with heart in the same sentence and with the minimum
frequency of 6, data was selected (English: 254, Korean: 614) and classified into one of the four
components.
Frequency by component (Figure 1) shows that the two cultures exhibit both similarity and
difference in their preference of highlighting components of the frame: Both cultures highlight the
feeling pain component most and the causing pain component next; Korean highlights the feeling pain
and wound components to higher degrees than English does, and English perspectivizes towards the
causing pain and the responding components more often than Korean does. Examining the
expressions of the components revealed that Korean has more diverse pain tokens than English does
and also has a distinctive wound token of ‘mot’ (nail), and that the English causing pain component
also connects with the additional component of bleeding (eg. My heart bleeds for her), which Korean
doesn’t. The results show Korean culture pays more attention to introspective responses to sadness
than English culture does, and the latter to more outward ones than Korean.
Components
Frame of Pain
Causing Pain
of
Feeling Pain
Responding to Pain
Wounds
English Example
70
Her voice was needle-sharp and her words
pierced Sophia’s heart.
I can taste the liquor in my throat, which I
drank to still the pain in my heart.’
The pain in Jake's face was making her heart
weep.
It would take much more to heal the wounds
to her heart.
<Table 1>
60
50
40
English
30
Korean
20
10
0
causing
wounds
<Figure 1>
References
Bartsch, Renate. 2002. Generating Polysemy: Metaphor and Metonymy, Metaphor and Metonymy in
Comparison and Contrast, R. Dirven and R. Pörings (eds.), Berlin: Mouton de Gruyter, 4974.
Köch, Peter. 2001. Metonymy: Unity in Diversity Source, Journal of Historical Pragmatics, Vol. 2.2,
201-244, John Benjamins Publishing Company.
Kövecses, Zoltan. 2000. Metaphor and Emotion: Language, Culture, and Body in Human Feeling,
New York: Cambridge University Press.
Talmy, Leonard. 1996. The windowing of attention in language. Grammatical Constructions: Their
Form and Meaning, Masayoshi S. & S. Thompson (ed.), 235-287. Oxford: Oxford University
Press.
*A unit consisted of a word and a postposition
What to wear today?
Convergence and divergence in Dutch clothing terminology
Jocelyne Daems
Quantitative Lexicology and Variational Linguistics (QLVL), KU Leuven
This paper reports on a corpus-based investigation into naming preferences in Belgian Dutch and
Netherlandic Dutch for fourteen clothing terms. The study (Daems et al., accepted) is a follow-up of
Geeraerts et al. (1999), in which soccer and clothing terminology from 1950, 1970 and 1990 was
analysed as an indicator of standardisation in Dutch. This study extends the clothing corpus with new,
comparable data from 2012 collected from magazines, to represent standard language, and shop
windows, as a substandard register. Focusing on lexical uniformity, we rely on the onomasiological
measure of lexical variation designed for the aforementioned study, which calculates the differences
in lexicalization preferences for a given concept in the two regions. For example, Table 1 shows the
concept OVERHEMD ‘(dress) shirt’, which can be lexicalized by hemd, overhemd and shirt. The degree
of uniformity between Belgian Dutch and Netherlandic Dutch can be measured in terms of
overlapping lexicalization preferences (summing the smallest relative value for each term: (19+0+7) =
26%).
OVERHEMD (MAN)
Neth.Dutch
Belg.Dutch
hemd
19%
93%
overhemd
12%
0%
shirt
69%
7%
Table 1 - Lexicalization preferences for OVERHEMD (MAN)
The present-day material is compared to the uniformity levels obtained by Geeraerts et al. (1999),
which shows new insights on the linguistic situation in the Low Countries. The diachronic
convergence between Belgian Dutch and Netherlandic Dutch confirmed in the original study seems to
have come to a halt in 2012. Furthermore, the recent data suggest that the distance between the
lower register shop windows and the standard language in magazines remains largest in Belgian
Dutch.
References
Daems, Jocelyne, Kris Heylen & Dirk Geeraerts. Accepted. ‘Wat dragen we vandaag: een hemd met
blazer of een shirt met jasje?’. [Accepted for publication in Taal en Tongval.]
Geeraerts, Dirk, Stefan Grondelaers & Dirk Speelman. 1999. Convergentie en divergentie in de
Nederlandse woordenschat. Een onderzoek naar kleding- en voetbaltermen. Amsterdam: P.J.
Meertens-Instituut.
Perception of Non-Native Errors: How Native Speakers of German Judge and Comprehend
Erroneous German Produced by Native Speakers of Russian
Evghenia Goltsev
University of Freiburg
Numerous studies are devoted to the issue of the perception and evaluation of non-native errors.
Many have attempted to establish a hierarchy in error gravity by referring to the deviations at different
linguistic levels, i.e. phonology, morpho-syntax, lexis. However, only a few have taken into account
the frequency aspect (Albrechtsen et al. 1980). The present project investigates the effect of errors on
native speakers’ perception, considering the interaction between different linguistic levels and
frequency, and with regard to salience. The frequency types of error occurrences analyzed within the
study are, e.g. density and frequency of use. In particular, these types can be defined as a certain
number of deviations of different kinds in one unit of text [density], and the frequency of the feature
containing the deviation in usage of native speakers [frequency of use] (James 1998). The main
hypotheses to be tested are that e.g. errors in high frequency items evoke a negative judgment in the
listener: however, they are less likely to impair comprehensibility than in low frequency items, since in
the top-down process of oral comprehension frequently occurring units are more predictable and
therefore more recognizable (Diessel 2007; Bybee 2007).
This study can be located within the usage-based approach drawing on the assumption that
actual language use and how it is processed can affect listener’s mental representation of language
(Ellis 2007). The ongoing data collection is conducted by means of sociolinguistic and psycholinguistic
methods (Garret 2010). Different audio stimuli representing the above-mentioned frequency and error
types are compiled using the verbal-guise technique (Giles / Powesland 1975). The stimuli are
presented to 200 native speakers of German who rate them according to the personal traits of the
speaker and the speech itself. The subjects also perform a multiple-choice comprehension task.
The stimuli were based on an analysis of a corpus of 35 half-hour interviews with Russian
speakers of German. The analysis revealed which errors they are most likely to make, for example
overproducing the pronoun “es” with verba dicendi. The error types were tested regarding their
salience in a web-based perceptional experiment with 65 participants before being applied to compile
the stimuli. Syntactical errors were shown to be quite salient, whereas some phonetic deviations
seem to be scarcely noticeable, e.g. in comparison to an absent glottal stop, the verb in 2nd position
received a significantly higher probability (p < 0.001) of being recognized. In order to determine the
quantitative constraints for the frequency types, a further perceptional test was undertaken with 50
subjects. This survey provided information on how different error ratios are perceived and processed,
i.e. 17% of erroneous tokens in one text are judged as a large error quantity leading to
comprehension difficulties. The average rating of a text with 17% error-token ratio was significantly
higher (p < 0.001) than the average rating of a text with 12% error-token ratio.
References
Bybee, J. 2007. Frequency of Use and the Organization of Language. New York
Diessel, H. 2007. Frequency Effects in Language Acquisition, Language Use, and Diachronic
Change. New Ideas in Psychology 25, 108–127.
Ellis, N. 2009. Optimizing the Input: Frequency and Sampling in Usage-based and Form-focussed
Learning. Long, M.H.; Doughty, J.C. (Ed.) The Handbook of Language Teaching, 139 – 158.
Garret, P. 2010. Attitudes to Language. Cambridge.
Giles, H.; Powesland P. 1975. Speech Style and Social Evaluation. Language in Society, 428-433.
James, C. 1998. Errors in Language Learning and Use: Exploring Error Analysis. London.
Dynamic processes in the syntax of Russian northern dialects
1
(the functioning of the particle conjunctions)
Elena R. Guseva
Petrozavodsk State University
In contrast to the standard language, which in the past two centuries the number of
conjunctions with differentiated meanings (for the reason that, по той причине что, etc.) has been
increased, in the dialect speech the conjunctions, which have simple structure, are stored and formed.
Some of them, particle conjunctions, are developed as a result of agglutination of the ancient
elements - particles *li, *le, *bo, *že and others under the influence of the phonetic law of
Wackernagel [1].
In this paper we consider structural and semantic characteristics of 2-5-component
clusters from the position of synchronous-diachronic approaches, their genetic relations and areal
distribution in the Slavic languages: the facts of Russian North dialects are compared with the data
historical, etymological, interpretive dictionaries, also the data Middle-, Southern and other Russian
dialects recorded in the dialect dictionaries are attracted to analyze.
Dialect conjunctions with n-, ž-particles (azhno, izhno, dazhe, aino, etc.) form semantic
group: basic meaning no-particle is adversative, že-particle are adversative and conclusive.
The study shows that two-component clusters like a+zhe and others, dating back to the
*aže, jže, *da že, *a ono or a no / nъ are Slavic; azhe, izhe, dazhe, ano are marked in the Old
Russian language. The conjunctions a/zh/no, i/zh/no, da/z/no, a/i/no and the like consisting of three
components are known in the Russian language, mostly in modern dialects. Wide areal distribution of
dialect conjunctions azhno, izhno, aino and izhne, izhnu, azhnu, azhne and others indicates their
antiquity. Perhaps they were used in the Old Russian language, in colloquial speech (they are not
recorded in the historical dictionaries of the Russian language, with the exception of azhno, which is
marked in the ancient written materials from the XV century. The presence of particle no in the
conjunctions, which almost is not observed in ancient sources as a standalone unit, is important;
apparently its fusion with two component clusters azhe, izhe, ai was in ancient times. T.M. Nikolaeva
indicates that the formation of lexemes was carried out to around XVI century by agglutination
method.
The Dictionary of the Russian language in the XVIII century notes that only vernacular
conjunctions dazhe, azhno were used at this time. The remaining conjunctions (2-3-component
clusters) are not recorded in the dictionary, hence they functioned in colloquial dialect speech.
In modern Russian the using dazhe, azhno are limited stylistically: azhno commonly used
in the illiterary colloquial speech, dazhe - in a literary colloquial speech. In the Modern dialect speech
dazhe and azhno are polisemantic: the range of meanings of the conjunction azhno are expanding
compared to the lexemes of the Old Russian language, the number of meanings of the conjunction
azhno are reduced; dialect conjunction izhno is polisemantic. The range of meanings of 3-component
dialect clusters is broader than the ancient 2-component units.
Not only three, but four-particlers clusters are used in dialect speech: a/zh/na/k, a/zh/ni/k,
a/zh/ni/t’, a/zh/no/i, i/zh/na(y)/k, i/zh/no(e)/t’. 4-component conjunctions are not fixed in the historical
dictionaries. They were formed in the later period and were used in dialect speech. Changes in the
semantic of conjunctions with the accession of the fourth component are not happening. The 5component cluster a/zh/no/li/cha is fixed in the end of XIX - the beginning XX century in the southern
dialects.
So, in dialect speech, including the dialects of the Russian North, there is preservation of
ancient etymologically unmotivated particularal lexemes. Dependence of semantic volume of lexems
are observed of their structure.
References
1 Николаева Т.М. Непарадигматическая лингвистика: (История блуждающих частиц). М., 2008.
1
. Работа выполнена при поддержке Российского гуманитарного научного фонда (РГНФ).
Проект № 15-04-00180.
Metaphors and blending in Erzya and Šokša Mordvin disease names
Flóra Hatvani
The Erzya Mordvin language belongs to the Finno-Volga branch of the Uralic language family. The
biggest Mordvin (Erzyas and Mokšas and Šokšas) communities live in Mordovia, but there is a
Mordvin diaspora in other regions of the Russian Federation. Within the Finno-Ugric language
communities the process of assimilation is advancing considerably and this holds especially for the
Mordvins. The diseases names are really good examples of how the original – and motivated –
expressions are disappearing from the spoken language.
In this talk I will discuss the Erzya and Šokša Mordvin disease names with respect to their
motivation in a cognitive semantic framework. I will also compare the Erzya data with Šokša Mordvin
examples. The data comes from my own fieldwork in 2013 and 2014.
Most of the Erzya disease names are motivated and often based on metaphors. The disease
names are largely dependent on the cultural background (venčamo porazo saś ’pimply’, lit. ’someone
who has to get married soon’), the enviroment (ŕiveźeń tol ‘chickenpox’, lit. ’fox’s fire’) and the
religion/beliefs in the Erzya language community.
Several disease names are based on conceptual metaphors, for example ľivťemka 'herpes,
rash' or pačkľivkst 'boil', which are instances of a BODY = CONTAINER metaphor type since they
denote the disease as coming out of the human body. The verbs also have this kind of orientation:
ľivťiťi ‘become pimply, will be a boil’ lit. ’come out’. The Hungarian disease names display the same
pattern: kiütés ’rash’, kinövés ’boil’. Nevertheless, the opposite direction is also found, but occurs only
in curses: „Aparo orma potmozo sovazo lit. ’bad disease would enter to his/her stomach’.
Some disease names are based on another conceptual metaphors: DIRTY/UGLY =
BAD/IMMORALITY for instance avańks orma ‘syphilis, venereal diseases’ lit. ‘dirty disease’, amazij
orma ‘heart attack, syphilis, venereal diseases’ lit. ‘ugly disease’; as well as DARK = BAD/DEATH for
instance raužo orma ‘fatal diseases lit. ‘black disease’. On the other hand there are euphemisms
which denote the same diseases: paro orma ‘fatal diseases’ lit. ‘good disease’, večkemań orma
(Šokša) ‘venereal diseases’ lit. ’love’s disease’. These metaphoric terms are interesting because most
of them are polysemic (for example, aparo orma ’cancer, syphilis, heart attack, paralysis, long lasting
fatal disease, sudden death disease, not curable disease’). However, it should be mentioned that the
Mordvin speakers usually use the Russian term if they want to specify the disease.
There are some disease names which can be analysed as instances of blending:
sarazmukorvaŕa ’wart’ < saraz ‘hen’ + mukor ’bottom’ + vaŕa ’hole’, sukaodar ’boil on the armpit’ <
suka ‘female’ + odar ‘udder’, pińeodar ’boil on the armpit’ < pińe ‘puppy’ + odar ‘udder’ (Šokša) and
pińeń čečej ‘stye’ < pińeń ‘puppy's’ + čečej ‘spleen’. In this case the new concept/disease name is a
combination of its elements. In my poster I will exemplify the conceptual integration of these disease
names with figures.
A matter of taste: Comparing Turkish, French, and English winery tasting notes
Christina Hostetler
Ball State University
Wine tasting, to those within the wine culture, is more than simply the sense of taste. In fact,
experiencing wine involves a fusion of inputs from different senses, combined with some
sophisticated processing in the brain. Sense in many respects is relative, so that even though a group
may experience the same wine, they may in fact experience from different perspectives. However, it
has been claimed that through practice, experts develop the ability to distinguish subtle yet distinct
aromas and flavors that novices or non-expert (Moore & Charling 1988).
Many descriptions of wine have been designed as guides for both novice and expert wine
enthusiasts, yet the literature has expanded in the last half a century to include observations of wine
from multiple perspectives, especially philosophy and cognition (Brochet & Dubourdieu. 2001). One of
these perspectives has been linguistic analysis of the language used to describe wine. Some studies
have focused on lexical choice in wine descriptions (Brochet & Dubourdieu. 2001, Lehrer. 1983).
Other studies have considered the
While in many ways, the research on wine descriptions is extensive, the current study sought
to compare wine descriptions from different cultures in the areas of lexical and rhetorical structures.
The study considers previous research studies such as the studies by Normand (1998) and Brochet &
Dubourdieu (2001) that have focused on the lexical items used in tasting notes in French and English.
By comparing French, English, and Turkish wine tasting descriptions, the current study adds to the
previous research by not only adding a different language and culture, but also adding the syntactic
and rhetorical structures. Data was collected from Turkish, French, and American wineries’ online
descriptions of their wine. Both a white and red wine varietal was collected from each winery and
translated into English. The translations were then analyzed individually using measures for rhetorical
and lexical structures previously used in similar studies then compared both quantitatively and
qualitatively. Emphasis was placed on the metaphorical meaning of the words.
The results show that while within the overarching wine culture, there are similarities in
individual lexical items and rhetorical structures, distinct metaphorical structures in each language
also surface. The universal rhetoric and lexicon of wine descriptions reflects the universal structures
established by those in the wine community that have developed an accepted identity of their own.
Whereas the unique metaphorical descriptions and subtle cultural specific references create
distinctions between the wine descriptions based on language and nation. The hypothesized reason
for these distinctions is the connection between wine production and cultural values.
References
Brochet, Frederic & Denis Dubourdieu. 2001. Wine descriptive language supports cognitive specificity
of chemical senses. Brain and Language. 77, 187-196.
Caballero, Rosario. 2007. Manner-of-motion verbs in wine description. Journal of Pragmatics, 39,
2095-2114.
Lehrer, Adrienne. 1983. Wine and Conversation. Bloomington, IN: Indiana University Press.
Moore, Terrance & Chris Carling. 1988. The Limitations of Language. New York, NY:
St. Martin’s Press.
Nedilko, Anatoly. 2006. Viticulture and winemaking terminology and terminography: A review of
resources. Terminology, 12(1), 137-164.
Normand, Sylvie. 1998/1999. La terminologie de la degustation du champagne. Terminology, 5(2),
251-269.
Pretty Little Chunks of Language: A usage-based approach to teaching formulaic
language to young L2 learners
Saskia Kersten
University of Hertfordshire
Following a study which showed that young learners in primary school benefit from learning English
as a second language based on cognitive linguistic theory (Kersten 2010), this poster provides a first
tentative outline of and invites feedback on a study that is designed to explore the role of formulaic
sequences (Wray 2002) in second language development from a usage-based perspective.
As studies have shown, L2 learners do use “conventionalised ways of saying things” (Smiskova et al.
2012), but not always in the same way native speakers would. Although native-like selection of multiword units is not necessarily the only goal of second language development, a good command and
extensive knowledge of formulaic language have been shown to aid, for example, fluency in the L2
(Wood 2010).
For example, some advocate that collocations should be explicitly taught, including explicit
comparison with the learners’ L1 (Nesselhauf 2003), while other state that awareness raising and
semantic and structural elaboration (Boers & Lindstromberg 2009) may be used to foster chunk
knowledge and use in learners.
Most studies and recommendations on teaching chunks of language focus on intermediate or
advanced learners and often include a period of immersion in the target language community or
bilingual education programmes (see Boers & Lindstromberg 2012 for a review of studies since
2004).
The aim of the proposed study is to investigate the possibility of fostering the development of
formulaic language in the early stages of second language development and in a non-immersion
setting. This has implications for the ways of teaching chunks of language, following methodologies
suitable for young learners. In addition, the way data are collected has to be adapted accordingly.
Suggestions for the above will be made and example of potential teaching material will be given.
References
Boers, F. and Lindstromberg, S. (2012). Experimental and Intervention Studies on Formulaic
Sequences in a Second Language. Annual Review of Applied Linguistics, 32, 83–110.
Boers, F. and Lindstromberg, S. (2009). Optimizing a Lexical Approach to Instructed Second
Language Learning. Basingstoke: Palgrave Macmillan.
Kersten, S. (2010). The Mental Lexicon and Vocabulary Learning: Implications for the foreign
language classroom. Tuebingen: Narr.
Nesselhauf, N. (2003). The Use of Collocations by Advanced Learners of English and Some
Implications for Teaching. Applied Linguistics, 24, 223–242.
Smiskova, H., Verspoor, M. and Lowie, W. (2012). Conventionalized ways of saying things (CWOSTs)
and L2 development. Dutch Journal of Applied Linguistics, 1, 125–142.
Wood, D. (2010). Formulaic Language and Second Language Speech Fluency: Background,
Evidence and Classroom Applications. London: Continuum.
Wray, A. (2002). Formulaic Language and the Lexicon. Cambridge: CUP.
Metaphorical Patterns of HEART AND HEAD in English and Chinese: a Corpus-based Study
LI, Yi
Shandong University of Finance and Economics, China
The embodiment thesis views the more abstract target domains of cognition (mind) as based on
concrete source domains such as the human body (parts). The study takes the interactive view of
embodiment and offers a corpus-based and comparative analysis of various conceptualizations of the
body (parts), the HEART and the HEAD, for inner life experiences in English and Chinese. The
primary aim is to reexamine the role of the body in abstract conceptualizations by exploring body and
culture as groundings of metaphor. The second aim is to address the critical methodological issues
about systematic identification of linguistic and conceptual metaphors in natural data from large-scale
corpora.
This study employs the Pragglejaz Metaphor Identification Procedure (MIP) for identifying linguistic
metaphors, and the WordNet and SUMO method for identifying conceptual structures of linguistic
metaphors. Two conceptual metaphors for mental life, HEAD/HEART AS A CONTAINER and
HEAD/HEART AS AN OBJECT OF MANIPULATION, are found both in English and in Chinese. A
difference seems to exist in terms of degree of saliency or conventionality of the shared conceptual
metaphors, with HEAD AS A CONTAINER in English and HEART AS AN OBJECT OF
MANIPULATION in Chinese as the most productive metaphor for mental life respectively. Both
English and Chinese share the conceptual metaphors HEART AS AN OBJECT, HEART AS A
CONTAINER, and HEART AS MOTION for emotional life. A difference seems to exist in terms of
degree of saliency (or conventionality) of the shared conceptual metaphors, with HEART AS AN
OBJECT in English and HEART AS A CONTAINER in Chinese as the most productive metaphor for
emotional life respectively. The difference in saliency of the shared conceptual metaphors is a
reflection of the relative preferential conceptualizations of body (parts) for inner life experiences in two
different cultures.
In addition, cross-cultural variations in metaphor are observable in the “range of the target” that the
two languages and cultures have available for the conceptualizations of the head and/or the heart.
English uses two source domains, while Chinese selects five source domains to characterize mental
life. HEART AS A LOCATION, HEART AS MOTION, and HEART AS LIGHT, are alternative
metaphors for mental life in Chinese. English employs three source domains, while Chinese selects
five source domains to characterize emotional life. HEART AS HEAT and HEART AS FOOD are
alternative metaphors for emotional life in the Chinese language. The underlying reason is that the
“range of the target domains” is extended in the Chinese language and culture, resulting in the crosscultural variations in English and Chinese.
The overall claim of the study is that metaphor is a widespread phenomenon that occurs in the
interactions of body, language, and culture. The methodological separation between identifying
linguistic and conceptual metaphors has a clear advantage over the top-down approach that would
start out from conceptual metaphors in the study of metaphor in language and thought.
Using Fables to Prompt Complex Speaking in Young Adults
Marilyn A. Nippold
University of Oregon
This study was designed to examine syntactic complexity in the spoken discourse of young adults.
Two speaking tasks were employed, conversational and narrative. It was predicted that the narrative
task would elicit greater syntactic complexity than the conversational task.
The participants were 40 adults (mean age = 22 years; range = 20-24 years) who demonstrated
typical language development, based on norm-referenced testing using the Test of Adolescent and
Adult Language-Fourth Edition (TOAL-4; Hammill, Brown, Larsen, & Wiederholt, 2007). All adults
were native speakers of General American English, living in the western United States.
Each participant was individually interviewed using a conversational task, followed by a narrative task.
The conversational task encouraged the participant to talk about common topics such as school,
work, family, friends, and travel. For the narrative task, the participant listened to two Greek fables,
retold each one in his or her own words, and answered a set of reflective questions about the stories.
Fables were used for the narrative task because they address complex topics about human beings,
and therefore should be of interest to young adults. For example, the fables used in the present study
– The Monkey and the Dolphin and The Mice in Council – prompted the participant to think about
complex mental states including goodwill, trust, pretense, disgust, fear, pride, wisdom, and sacrifice.
After the participant retold a fable, the examiner asked questions that prompted reflection on the
deeper meanings of the story. For example, for The Monkey and the Dolphin, the participant was
asked to draw inferences (e.g., “Why did the dolphin ask the monkey if he was an Athenian?”),
interpret an idiom (e.g., “Can you explain the meaning of the expression, to be in deep water?”), and
think critically about the moral message it conveyed (e.g., “Do you agree with the moral of this story,
that those who pretend to be what they are not, sooner or later, find themselves in deep water?”). It
was expected that these questions, in addition to the story retell, would prompt the use of complex
syntax. Syntactic complexity was measured by calculating mean length of C-unit (MLCU) and clausal
density (CD) for each speaking task.
As predicted, MLCU was greater on the narrative task (mean = 11.89 words; range = 8.78-17.27
words; SD = 1.95) than on the conversational task (mean = 9.58 words; range = 5.64-13.44 words;
SD = 1.61), a statistically significant difference (t = 7.54, p < .0001, d = 1.30) with a very large effect
size. Additionally, CD was greater on the narrative task (mean = 2.28; range = 1.67-3.25; SD = 0.34)
than on the conversational task (mean = 1.68; range = 1.05-2.25; SD = 0.26), also a statistically
significant difference (t = 10.63, p < .0001, d = 2.0) with a very large effect size.
Syntactic complexity was especially high in response to the reflective questions. For example, the
following C-unit, which contains 35 words and seven subordinate clauses, was produced by a 22year-old adult as she explained why she agreed with the moral of the fable, The Monkey and the
Dolphin:
And you know, I’ve actually heard stories about people who try to do things like lie about
taking a sabbatical and take a job somewhere else and then it comes out in the newspaper.
The key message of this study is that tasks that prompt reflection on complex topics are more likely to
reveal a young adult’s capacity to use complex syntax than are more traditional conversational tasks.
Constructicon Building as a Practical Implementation of Construction Grammar and Frame
Semantics
Kyoko Hirose Ohara
Keio University
This paper gives an overview of the Constructicon-building effort within the Japanese FrameNet
project, viewing the project as a practical implementation of the theories of Construction Grammar
and Frame Semantics (cf. Boas 2010, Fillmore & Baker 2010, Fillmore et al. 2012). The paper argues
that by understanding a Constructicon-building effort within a FrameNet-building project as a practical
implementation of Construction Grammar and Frame Semantics, it becomes possible to define a
division of labor between frame annotation and construction annotation. The paper furthermore
proposes a new classification of grammatical constructions within a Constructicon, focusing on
relations between grammatical constructions, “semantic” frames and “interactional” frames and on an
analysis of Japanese constructions.
The Berkeley FrameNet project and its sister projects in Japanese, Swedish, and Brazilian
Portuguese are describing the semantic and distributional properties of words, based on Frame
Semantics. They have started to build the Constructicon, the registry of constructions in each
language, as well, in order to describe the meaning of sentences as a whole. However, how to relate
frames and constructions in these projects have been problematic. Following Ohara (2014a, b), this
paper points out that as practical implementations of Frame Semantics and Construction Grammar,
the syntax-lexicon continuum must be preserved in the FrameNets and the Constructicons. Thus,
lexical units in the FrameNets and the constructions in the Constructicons should be kept parallel.
Furthermore, the paper proposes that we can define: frame annotation as pertaining to frame-based
meaning structures of lexical units and constructions; and construction annotation as consisting of
syntactic annotation and frame-based and non-frame based meaning-structure annotation based on
semantic and interactional frames.
The present paper proposes a new five-way classification of constructions, based on whether
constructions evoke “semantic” frames or not and, additionally, whether “interactional” frames are
involved or not. Whereas “frames” in FrameNets are “semantic” frames, that is, “script-like conceptual
structures that describe a particular type of situation, object, or event along with its participants and
props” (Ruppenhofer et al. 2010:5), “interactional” frames are meaning structures which cannot be
characterized as consisting of participants and props (i.e., frame elements) but which are negotiated
in interaction between the context and the speaker.
To summarize, by viewing Constructicons as implementations of Construction Grammar and
Frame Semantics and thus by maintaining the syntax-lexicon continuum in Constructicons, it is
possible to clarify the relations between constructions and frames. The paper argues that
constructions can be classified taking into account not only “semantic” frames described in
FrameNets, but also “interactional” frames, which are integral part of many constructions, at least in
Japanese.
Selected References
Boas, Hans C. (2010). “The Syntax-lexicon continuum in Construction Grammar”. Belgian Journal of
Linguistics. 24, 54-82.
Fillmore, Charles J. and Collin F. Baker (2009) “A Frames Approach to Semantic Analysis”. In Heine,
Bernd and Heiko Narrog (Eds.), The Oxford Handbook of Linguistic Analysis, 313-339. Oxford:
Oxford University Press.
Fillmore, Charles J., Russell R. Lee-Goldman, and Russell Rhodes. (2012). “The FrameNet
Constructicon”. In Boas, Hans C. and Sag, Ivan A. (Eds.), Sign-based Construction Grammar,
309-372. Stanford: Center for the Study of Language and Information.
Ohara, Kyoko Hirose. (2014a). Relating Frames and Constructions in Japanese FrameNet.
Proceedings of the Ninth International Conference on Language Resources and Evaluation
(LREC '14), 2474-2477.
Ohara, Kyoko Hirose (2014b). “Relations between Frames and Constructions in Japanese FrameNet:
th
The Syntax-Lexicon Continuum Revisited”, 8 International Conference on Construction Grammar
(ICCG8)
Ruppenhofer, J., Ellsworth, M., Petruck, M.R.L., Johnson, C.R. and Scheffczyk, J. (2010). FrameNet
II: Extended Theory and Practice. Berkeley, California: International Computer Science Institute.
Beat-like gestures use in different types of speech in American English.
Mathilde Peyré
Lille 3 University
In studying multimodal communication, we focus on the way speech and gesture work together to
create and convey a message. Parallel to the referential content, there is also a pragmatic aspect to
the use of gesture which is to assist and regulate the delivery of this content. Beat-like gestures serve
such a pragmatic purpose and have received relatively little attention in comparison to referential
gestures. The present study concerns the use of beat-like gestures in different types of discourse in
English; with beat-like gestures we mean any type of rhythmic body movement used to punctuate and
reinforce speech. This study will focus on hand, head and eyebrow movements. Numerous studies
have been conducted on the semantic relationship between speech and iconic gestures (McNeill
1992, 2000, Kendon 2004, Goldin-Meadow 2003), but few empiric studies (apart from McClave 1991,
Loehr 2004, Loehr 2013) address the interaction between speech and beat-like gestures. McClave
and Loehr found that the rhythmic up and down movements (called beats) tend to align with the
nucleus of the intonation units.
This paper presents a gestural analysis of four types of discourse: political speeches, TV debates,
teaching and spontaneous conversation. These four types could be seen as forming a cline from the
most controlled type to the most spontaneous. For each type, 7 video clips of 1 minute each have
been selected, representing altogether 1031 gestures for 28 minutes of video. The type of gesture,
the alignment with accented syllables, as well as the amplitude and the intensity of the movement
were annotated in ELAN. Amplitude was calculated thanks to a program created with the software
MATLAB. The gestures were then put into three categories: small, medium and large. The annotation
of intensity is based on the speed and amplitude of the movement.
Given the more public and emphatic nature of the controlled types of discourse in our data set, one
expects more numerous, larger, more intense beat-like gestures and greater alignment with stressed
syllables in more controlled speech.
The results confirm that, for all discourse types, beat-like gestures tend to align with the accented
syllables in the corresponding speech, even though the alignment is not systematic. There is no
evidence in the data that there is more alignment when the type of discourse is more controlled. Also,
the results do not confirm the expectation of a lower frequency of beat-like gestures along the
controlled-speech hierarchy suggested above (ranging from political speeches over TV debates to
teaching). However, the data show a clear discrepancy between these three types of discourse on the
one hand, and spontaneous conversation, on the other, in terms of quantity, size, and intensity of
beat-like gestures. First of all, spontaneous conversation presents a lower total number of beat-like
gestures. Secondly, speakers in all four types of discourse tend to use small gestures, yet the amount
of large gestures is much lower in spontaneous conversation. The amount of high intensity beat-like
gestures also appears to be lower in spontaneous conversation than in the other types of discourse.
In sum, there seems to be a difference between the way speakers use beat-like gestures when in
spontaneous conversation on the one hand and in the three other types of discourse on the other,
however it is not clear whether this is connected to the more or less formal nature of discourse. The
next step for this study is to investigate the different functions of the types of discourse. Indeed, the
differences could be linked to the more narrative function of discourse in spontaneous conversation
and the more explicative and argumentative functions in the three other types of discourse.
References
Golding-Meadow, S. (2003). Hearing gesture: How our hands help us think, Cambridge MA: Harvard
University Press
Kendon, A. (2004). Gesture, Visible Action as Utterance, Cambridge University Press.
Loehr, D. (2004). Gesture and Intonation. Doctoral Dissertation, Georgetown University, Washington,
DC.
Loehr, D. (2013). Gesture and Prosody. in Müller, C., A. Cienki, E. Fricke, S.H. Ladewig, D. McNeill &
J. Bressem (eds.) Body – Language – Communication. An International Handbook on
Multimodality in Human Interaction. Berlin/Boston: De Gruyter Mouton.
McClave, E. (1991). Intonation and gesture. Doctoral Dissertation, Georgetown University,
Washington, DC.
McNeill, D. (1992). Hand and Mind: What Gestures Reveal about Thought, Chicago University Press.
McNeill, D. (2000). Analogic/Analytic representations and cross-linguistic differences in thinking for
speaking. Cognitive Linguistics 11, 43-60.
Corpus Based Grammatical Order for Learning Contents Outline of Thai as a Foreign
Language (Part 3: Modality markers)
Watit Pumyoo
Chiang Mai University, Thailand
This present research is a study of the frequency of modality marker usage in Thai based on
attested language data, the Thai National Corpus. It aims at analyzing the frequency of usage and
variant grammatical functions of five modality markers, namely, “tong”, “khuan”, “naa”, “khong”, and
“aat”. Specifically, this study identifies the frequency of occurrence as modality markers and the
frequency of co-occurring patterns of the five modality markers in order to arrange the learning
contents outline of grammatical topics on modality in Thai as a foreign language (TFL). The modality
markers in Thai can be categorized into deontic modality markers (high-obligation “tong” and midobligation “khuan” / “naa”) and epistemic modality markers (high-conjecture “tong”, mid-conjecture
“naa”, and low-conjecture “khong” / “aat”). These markers are versatile and can be used as auxiliaries
preceding main verbs in sentences. It is found that their frequencies of occurrence as modality
markers can be ordered from highest to lowest as follows; “tong” (0.96), “khong” (0.95), “khuan”
(0.86), “aat” (0.85), and “naa” (0.40); and the frequency of the modality markers co-occurrence can be
ordered from highest to lowest as follows; “tong” (1.00), “khong” (0.20), “khuan” (0.20), “aat” (0.20),
and “naa” (0.20). From the frequency and cognitive syntactic/semantic analysis, “tong” is the most
prototypical and “naa” is the most peripheral. The five modality markers are plotted into image
schema and arranged in the TFL learning contents of a 150-hour-course per the following: “tong”
introduced in 75th hour, “khuan” introduced in 110th hour, “aat” introduced in 120th hour, “khong”
introduced in 125th hour, and “naa” introduced in 140th hour. From this arrangement of the modality
markers, a L2 learner acquires concept of modality from centre to boundary.
References
Bybee, J., R. Perkins, and W. Pagliuca. 1994. The Evolution of Grammar: Tense, Aspect, and
Modality in the Languages of the World. Chicago: University of Chicago Press.
Iwasaki, Shoichi and Ingkaphirom, Preeya. 2009. A reference grammar of Thai. Cambridge:
Cambridge University Press.
Langacker, R. 1991. Concept, Image, and Symbol. The Cognitive Basis of Grammar. Berlin and New
York: Mouton de Gruyter.
O’Keeffe, Anne, McCarthy, Michael, and Carter, Ronald. 2007. From corpus to classroom: language
use and language teaching. Cambridge: Cambridge University Press.
Phatranawig, Phornthip. 1972. Modal expression in the Thai language, unpublished licentiate thesis.
Bangkok: Chulalongkorn University.
Portner, Paul. 2010. Modality. Oxford: Oxford University Press.
Teubert, Wolfgang and Cermakova, Anna. 2007. Corpus linguistics: a short introduction. London:
Continuum.
Spatial opposition vs. visual accessibility and contrast: an experiment with Estonian
demonstratives
Maria Reile
University of Tartu
Demonstratives are deictic expressions that help to identify the referent to the addressee in linguistic
(endophoric use) or physical (exphoric use) context (Halliday and Hasan 1967). It is a widely held
view that exophoric demonstratives are differentiated mainly on the basis of the location of the
referent. In recent years there has been an increase of studies on the use of exophoric
demonstratives and factors that affect the choice between proximal and distal demonstratives but
mostly in Indo-European languages (e. g English, Dutch) and with contradictory results. (e.g Piwek et
al 2008, Jarbou 2010). To the author’s knowledge, the use of exophoric demonstratives in Estonian
has had little or no research attention. According to Pajusalu (2009), there are two possible
demonstratives in Standard Estonian: see ‘this’ and too ‘that’ but distal too is used mostly in SouthEstonian.
The aim of the study is to test the affective factors that are believed to have an effect on the
choice of exophoric demonstratives and to clarify the consistency of the exophoric use of distal too.
The hypothesis of the study is that the choice between demonstratives is not affected only by the
location of the referent but by visual accessibility of the referent and need for contrast as well.
To test the affective factors which are associated with demonstrative selection, an experiment
was designed based on earlier research (e.g. Diessel 2006, Piwek et al 2008, Coventry et al 2008,
Jarbou 2010). For eliciting data, stimuli of spatial opposition, visual saliency and contrast was used.
Selection and usage of demonstratives was explored in an artificial interactional situation consisting of
pairwise building of pre-determined figures out of Lego blocks. The Lego blocks were placed on a
varying distance on a large table in front of the participants. Thus, enabling to identify whether the
choice of the referent is affected by spatial opposition or other stimuli. Respondents were randomly
allocated into pairs of instructors and builders. 24 pairs of 16-19 year old students from three
secondary schools from different regions of South-Estonia were enrolled in the study. Video
recordings were transcribed and coded by the author. Descriptive statistics and qualitative content
analyses were used to interpret the findings.
Preliminary results show that to indicate spatial opposition of referents, participants use
proximal see ‘this’ for near referents and distal too ‘that’ for distant referents. For visually not salient
referents, participants tended to use proximal see ‘this’ (even if the referent was distant), rather than
distal too ‘that’, adding demonstrative proadverbs for location. To contrast the referents, participants
used proximal demonstrative see ‘this’ and pronoun teine ‘other’. Thus, preliminary analysis suggests
that not only the stimuli of spatial opposition play a role in selection of the demonstratives, but visual
accessibility of the referent and need for contrast as well. The experimental approach used in this
study proved to be effective in collecting linguistic data on spatial demonstratives and could be used
for further comparative research among other Finno-Ugric languages and between Finno-Ugric and
Indo-Germanic languages.
References
Diessel, Holger 2006. Demonstratives, joint attention, and the emergence of grammar. Cognitive
Linguistics, Volume 17, Issue 4, pp. 463–489.
Coventry, Kenny R., Bernice Valdés, Alejandro Castillo, Pedro Guijarro-Fuentes 2008. Language
within your reach: Near-far perceptual space and spatial demonstratives. – Cognition, Volume 108,
Issue 3, pp. 889-895.
Halliday, M. A. K, Ruqaiya Hasan 1976. Cohesion in English. Longman, London.
Jarbou, Samir Omar 2010. Accessibility vs. physical proximity: An analysis of exophoric
demonstrative practice in Spoken Jordanian Arabic. Journal of Pragmatics, Volume 42, Issue 11, pp.
3078–3097.
Pajusalu, Renate 2009. Pronouns and reference in Estonian. Sprachtypologie und
Universalienforschung, 62 (1/2), 122-139.
Piwek, Paul, Robbert-Jan Beun, Anita Cremers 2008. ’Proximal’ and ’distal’ in language and
cognition: Evidence from deictic demonstratives in Dutch. Journal of Pragmatics, vol 40, Issue 4, pp.
694–718.
Analysis of Evaluative Meanings of Conceptual Metaphors：Based on Semantic Tagging of
Wmatrix
Ya Sun
University of International Business and Economics, Beijing
Through the case study of metaphorical construction of the concepts of BUSINESS and
ENTERPRISE, this paper integrates corpus linguistics and the evaluation theory to provide a
framework for working out the evaluative meanings of conceptual metaphors with the help of the
online corpus tool Wmatrix.
Firstly, metaphor vehicles are identified with the help of MIPVU from the collocates of targetdomain words in the self-built corpus of business discourse so that conceptual metaphors used to talk
about the concepts under study can be found directly. Secondly, more than one semantic tag, which
is allocated automatically to each metaphor vehicle by Wmatrix, can be divided into two types. One
type of semantic tag, based on the definition of the source domain of conceptual metaphors, can be
regarded as the labels of source domains for metaphor vehicles in order to formulate conceptual
metaphors. The type of semantic tag, based on the definition of the evaluation of the appraisal
system, can be recognized as the specific attitudinal meanings of metaphor vehicles. Thirdly, the
specific attitudinal meanings serve as triggers to the evaluative meanings in the appraisal system and
thus the general semantic prosody for the corresponding conceptual metaphors.
It is found that the concepts under study are evaluated positively in terms of valuation and
capacity, the evaluative meanings of conceptual metaphors can be understood as the more
operational concept “register prosody”, and that the annotation of register prosody facilitates the
emergence of the interface among corpus methods, metaphor analysis and discourse analysis.
Representational distinctions between active-voice, be-passive, and get-passive
Dominic Thompson
University of Nottingham
Active-voice and passive-voice are complementary sentence forms that are available when describing
a transitive event. In English, the latter has two variants: be-passive and get-passive. Numerous
attempts have been made in the literature to represent the syntactic and semantic differences
between these forms, while maintaining their shared features (e.g. Fleisher, 2008; Haegeman, 1985).
Theories have often been overly complex, with little agreement between accounts. However, the pvP
Theory (Thompson & Scheepers, 2013) postulates a simple syntactic unit consisting of a “pv-Phrase”
with an unpronounced head, and a PRO specifier. This pvP unit is shared by both be-passives and
get-passives. The authors claim that this allows a clear representation of both the syntax and
semantics of these two passive forms.
The pvP states that the get- and be-passive differ in one representational aspect (number of
lexical verbs), and the be-passive differs from active-voice in one aspect (presence of pvP unit), while
the get-passive differs from the active-voice in both of these aspects. Following this logic, we suggest
that representational similarity will determine the pattern of changes in a paraphrasing task; that is,
paraphrases requiring a change to only one aspect will be preferred over those requiring two. This
interpretation is compatible with a ‘noisy-channel’ model of language comprehension (e.g. Gibson,
Bergen, and Piantadosi, 2013). The pvP also states that the broader range of meanings present in get
and absent in be will facilitate changes from get-passive to be-passive, rather than the other direction.
To test these predictions, we ran two experiments in which participants gave acceptability ratings for
various passive sentences and were given the option of providing an ‘improved’ free paraphrase for
each. In experiment 1, we manipulated passive-type (get vs be) and by-phrase (present vs absent),
as in (1).
(1)
The composer [was / got] seduced by the dancer
The composer [was / got] seduced
For experiment 2, in addition to the agentive by-phrase, we examining two further adjunct types (2).
(2)
The composer [was / got] seduced by the dancer
The composer [was / got] seduced by the end of the session
The composer [was / got] seduced in a dark back lane
In both experiments we found that be-passives were rated as more acceptable overall than getpassives, and were also less likely to receive an improved paraphrase.
As predicted, be-passives were almost exclusively changed into active-voice, while getpassives were significantly more likely to be changed into be-passives than into active-voice. This
pattern was present across both experiments and all adjunct types. The presence of an agentive byphrase increased the frequency of changes into active-voice, though primarily for be-passives.
Most significantly, despite active-voice being the most acceptable form, get-passives are still
preferentially changed into be-passives. This supports our prediction that it is the representational
similarity - number of representational differences between forms - that constrains the ease of making
a change: a change from get-passive to be-passive, or from be-passive to active-voice, requires only
one attribute to be modified. Although a change from get-passive to active-voice would result in
achieving the most canonical or acceptable form, such a change is more difficult, since two attributes
must be modified.
If only surface elements were relevant when quantifying the differences between sentences,
then changes either from be-passive to active-voice or from get-passive to active-voice would entail
the same amount of processing, and therefore should be equally likely to occur. Since this was not
the case, we suggests that the computational logic of noisy-channel models may extend beyond the
surface structure of a sentence to unpronounced structural elements, including the pvP unit.
References
Fleisher, N. (2008). Passive get, Causative get, and the Phrasehood of Passive vP. Proceedings of
CLS 41, 59-67
Gibson, E., Bergen, L., & Piantadosi, S. T. (2013). Rational integration of noisy evidence and prior
semantic expectations in sentence interpretation. Proceedings of the National Academy of Sciences
of the United States of America, 110(20), 8051-6
Haegeman, L. (1985). The get-passive and Burzio’s Generalization. Lingua 66, 53-77
Thompson, D. & Scheepers, C. (2013). Harmonizing the Passive: A new proposal for Passive
Constructions in Generative Grammar. Newcastle Working Papers in Linguistics 19.2
Representing irony in digital communication
1
2
Dominic Thompson , Hartmut Leuthold , and Ruth Filik
1
2
University of Nottingham; University of Tuebingen
1
In comparison to spoken language, computer mediated communication (CMC), such as text
messaging, lacks non-verbal cues that aid successful communication, for example, by minimising
ambiguity. It is therefore important to consider the strategies of language users in CMC to
compensate for this (Tossell et al., 2012). Irony is frequently used to communicate the opposite of
what is said: if someone makes a boring suggestion, their friend might say, “Sounds exciting!” This
opposition can give rise to ambiguity, making irony especially difficult to interpret in CMC.
A number of studies have examined the use of devices such as emoticons or punctuation to
aid understanding in CMC. However, most studies have involved participants rating the
appropriateness of a range of emoticons in certain contexts, or rating their emotional effect. There are
currently few production studies examining which emoticons are used in given contexts, and it
remains unclear which devices are used as indicators of irony, due to vague or conflicting findings
(Carvalha & Sarmento, 2009; Derks, Bos, & von Grumbkow, 2007). Furthermore, much of the
literature refers to irony generally, without distinguishing different types (e.g., criticism vs. praise). We
address this in two experiments investigating emoticon production in ironic versus literal contexts, as
well as in criticism and praise.
In Experiment 1, participants were presented with a document containing 48 short text message
conversations (e.g., 1a,b) in which the final comment was ambiguous between a literal or ironic
interpretation. They were asked to imagine these as conversations between themselves and a friend.
Their task was to clarify their intentions as either being serious (i.e., literal) or ironic, without adding or
removing words, with no explicit prompt to use emoticons. Filler trials were also included. The final
comment was also either superficially positive (1a) or negative (1b). When used in an ironic context,
the superficial polarity becomes reversed; that is, when intended ironically (1a) is interpreted as
criticism, while (1b) is interpreted as praise.
(1)
You:
Are you busy later?
Friend: Yes, I’m going to the guitar rehearsal for Sunday.
You:
(a) But you’re basically an expert already
you really need the practice
(b)
Yeah,
Participants produced an extensive range of devices: more than 90 distinct emoticons. In line with
several earlier studies (e.g. Carvalha & Sarmento, 2009), we found the most frequently used devices
in ironic contexts were ..., :p and ;). More interestingly, the use of each was modulated by the polarity
of the comment.
For ironic criticism, ... was the most frequently used device, and was used significantly more
frequently than in ironic praise. The ... device was also the most frequently used in contexts of literal
criticism. For ironic praise, :p was the most frequently used device, followed by ;), with both being
used more frequently than in ironic criticism. Likewise, :p and ;) respectively were the most frequent
devices in joking contexts, even displaying raw frequencies comparable to ironic praise.
In Experiment 2, we adapted the above design, removing the final comment and asking
participants to provide a response that would clearly communicate their intentions as being ironic or
serious. Again, there was no prompt to use emoticons. Despite allowing a more open-ended
response, we found that emoticon use was almost as high as in Experiment 1, and that the usage
patterns observed in Experiment 1 were replicated.
Of the huge variety of emoticons in use, a small set are used in facilitating the communication of ironic
intent. The devices in this set are not used uniformly, instead they vary with the polarity of irony.
Interestingly, the same device is used for both literal and ironic criticism; this suggests that ... is in fact
a marker of criticism. It may be that case that ... marks out criticism indirectly, with participants using
the device to ‘soften the blow’ of a critical comment. Nonetheless, use of ... clarifies that a superficially
positive comment is intended negatively (i.e. ironically). Conversely, the most popular device in ironic
praise ( :p ) is also the most popular when joking. If :p marks a joke, then participants may utilise it
when to indicate that a superficially negative comment should not be taken seriously.
Irony should not be conceptualised as simply inverting the meaning of a sentence; its precise
function differs depending on the given context. When irony is used for praise, it inverts the superficial
literal meaning, while aiming to be light-hearted, and hence less serious (or sincere) than literal
praise. This is in line with the Tinge Hypothesis, which predicts that irony lessens a comment’s
emotional impact (Dews, & Winner, 1995). Meanwhile, when irony is used for criticism, it inverts the
superficial literal meaning, but still retains the intention to be critical.
References
Carvalho, P., Sarmento, L., Silva, M. J., & Oliveira, E. de. (2009). Clues for detecting irony in usergenerated contents: oh...!! itʼs so easy;-). Proceeding of the 1st international CIKM workshop on
Topic-sentiment analysis for mass opinion, 53–56
Derks, D., Bos, A. E. R., & Grumbkow, J. von. (2007). Emoticons and Online Message Interpretation.
Social Science Computer Review
Dews, S., & Winner, E. (1995). Irony Muting the Meaning : A Social Function of Irony. Metaphor and
Symbolic Activity, 10(1), 3-19.
Tossell, C. C., Kortum, P., Shepard, C., Barg-Walkow, L. H., Rahmati, A., & Zhong, L. (2012). A
longitudinal study of emoticon use in text messaging from smartphones. Computers in Human
Behavior, 28(2), 659-663.
Conceptual Extension from Trajector to Landmark:
2
A Cognitive-based approach to SVC
LI Tianyu, Andrew Spencer
University of Essex
Serial verb constructions (SVCs) play an important role in typological studies, but there is still no
widely accepted definition of the construction (Paul, 2004). Aikhenvald (2006) proposes a definition
from the perspective of form and semantics, but this sometimes poses analytical problems when
distinguishing SVCs from other multiverb constructions.
This study argues that it is hard to define the notion of SVC consistently using traditional
grammar-based descriptions as the basis for definition, because of considerable cross-linguistic
variation in surface grammatical constructions. The authors therefore propose to apply a universal
conceptual perspective. A cognitive-based analysis of multi-verb constructions (most of which are
SVCs) in Mandarin Chinese has been carried out by Yin (2010), which made use of cognitive
grammar but did not go further to propose a precise definition. Besides, a cognitively based approach
to SVCs is less plausible if limited to the analysis of a single language.
In order to plug these research gaps, this study analyzes SVC from a cognitive-based
perspective (Langacker, 2003). It treats SVC as a kind of conceptual extension from the trajector to
the landmark within one event in certain languages that use two or more than two individual verbs to
encode this extensional process. This definition works well for the traditionally labeled asymmetrical
and symmetrical SVCs, in that the landmark of the former type is a static termination while landmark
of the latter is dynamic with a potential of further extension. In fact realization of that potential
extension would generalize multi events represented by other multiverb constructions (coordinate,
subordinate, etc.), therefore this cognitive-based definition can also locate SVCs among the
continuum of multiverb constructions.
The authors have tested this cognitive-based definition cross-linguistically on asymmetrical
and symmetrical SVCs along with their detailed subtypes. This provides a firm empirical base on
which to develop the proposals further.
References
Aikhenvald, Alexandra Y. & R. M. W. Dixon (eds.). 2006. Serial verb constructions: A cross-linguistic
typology. Oxford, U.K. & New York: Oxford University Press.
Langacker, Ronald W. 2003. Constructional integration, grammaticalization, and serial verb
constructions. Language and Linguistics, 4, 2: 251-278.
PAUL, Waltraud. 2004. The ―serial verb construction‖ in Chinese: A Gordian knot. La notion de
«construction verbale en série» est-elle opératoire? Fédération TUL Atelier du 9 décembre
2004, 1-23. Ehess, Paris.
Yin, Hui. 2010b. A Cognitive Approach to Multi-Verb Constructions in Mandarin Chinese. Unpublished
PhD dissertation. Canada: University of Alberta.
2
It is claimed that China Scholarship Council funds this research.
A corpus-based approach to causative-passive ambiguities in the get+NP+past
participle
Katsuko Tomotsugu
Shizuoka Institute of Science and Technology
Since Lakoff (1971) pointed out the ambiguous readings of John got his dishes washed, the
get+NP+past participle has been discussed in relatively few studies (Ikegami 1989; Gilquin 2010;
Emonds 2013) in contrast to the large amount of attention paid to the get+past participle (getpassives). The aim of this paper is to provide corpus evidence for the semantic factors involved in the
“transitive” get-passive gradient based on the British National Corpus. The basic premise is that the
verb groups of the concurrent past participles are not random, and the semantics of verbs in high
frequency comprises parts of the semantics of the get+NP+past participle.
From the BNC, 4528 samples of the get+NP+past participle were retrieved and grouped into
the five subclasses: disjoint reference (29.5%), active (co-reference) (54.7%), adjectival (10.9%),
psychological (2.5%), and formulaic (2.4%) (cf. five subclasses of get-passives in Collins (1996)).
The disjoint reference, which has three participants to fill in the subject, the object, and
(optional) by agents, is associated with the three semantic verb groups from the keywords of the past
participles: service situations (haircut, medical treatment, purchase, delivery, repairs), seeking
authoritative decisions (accept, approve, hear, sign, release), and violent situations (catch, kill, bite).
The first two groups yield the indirect causative, with the subject’s stronger control over the event and
the subject being positively affected from the event. The third group, with negatively affected subjects
and their possessive pronouns in the object, induces the indirect passive.
The active (co-reference), which has the identical subject and agent, allows for transitive
rephrasing using the two participants (you get it sorted out vs. you sort it out). The three dominant
semantic verb groups of the past participles are daily activities (do, set up, wash, write), starting or
finishing (start, finish), and solving problems (sort out, work out, clear up, organise, settle).
Completion of the event is often emphasized by the particles up, out, and away. Together with the
verbs of effort (try to, manage to) or modal verbs (have to, need to), get and the focused endpoint of
the event can be perfectly named as the successful accomplishment (the term used for have Object
pp. by Ikegami (1986)).
Two parameters resolve the ambiguous readings of the get+NP+past participle: three or two
participants and the degree of the subject control over the event. When the subject has less control,
the past participle is read passively in the disjoint reference and adjectivally in the co-reference.
References
Collins, Peter C. 1996. Get-passives in English. World Englishes 15(1). 43-56.
Emonds, Joseph. 2013. Indirect passives and the selection of English participles. Lingua 125. 58-75.
Gilquin, Gaëtanelle. 2010. Corpus, cognition and causative constructions. Amsterdam: John
Benjamins.
Ikegami, Yoshihiko. 1986. The drift toward agentivity and the development of the perfective use of
have + pp. in English. In Dieter Kastovsky & Aleksander Szwedek (eds.), Linguistics across
historical and geographical boundaries, Vol. 1, 381-386. Berlin: Mouton de Gruyter.
Ikegami, Yoshihiko. 1989. 'HAVE + object + past participle' and 'GET + object + past participle' in the
SEU Corpus. In Udo Fries & Martin Heusser (eds.), Meaning and beyond: Ernst Leisi zum 70.
Geburtstag, 197-213. Tübingen: Gunter Narr.
Lakoff, Robin. 1971. Passive resistance. Papers from the seventh regional meeting of the Chicago
Linguistic Society, 149-162.
Washio, Ryuichi. 1993. When causatives mean passive: A cross-linguistic perspective. Journal of
East Asian Linguistics 2(1). 45-90.
Pedagogical Conceptualization of English Articles
Benjamin White
Marshall University
ESL/EFL textbooks traditionally present multiple rules for article use followed by fill-in-the-blank
exercises. Relying exclusively on this approach is inadequate. First, exceptions to textbook rules are
easily found in authentic discourse (Pica, 1983). Second, rules may presume learners’ ability to make
difficult semantic distinctions not made in their first languages, such as the count/noncount distinction.
Third, individual rules often seem arbitrary and unrelated. For example, why do some textbooks claim
that no article (Ø) is used with both noun phrases that are the least definite and noun phrases that are
the most definite? Why use THE with names of rivers but Ø for names of lakes? How could this rule
for river and lake names possibly be related to the rule to use A for first mention and THE for
subsequent mention?
Following Negueruela’s (2009) call for pedagogical grammars to explain language through meaning
categories rather than rules, this poster details a conceptual framework for the English article system.
The framework prompts teachers and learners to view the grammar of English articles as a series of
conceptualizations to be experienced, not as a series of rules to be memorized. This approach falls
squarely within a concept-based approach to language instruction (Lantolf & Thorne, 2006).
The framework relies heavily on nominal grounding (Langacker, 2008), frame semantics (Fillmore,
2007), and metaphor in grammar (Steen, 2007). At its core, the framework provides a three-way
distinction: schematic images symbolize abstract conceptions for noun phrases headed by THE,
A/SOME, and Ø. These images place a noun phrase within a comparison set and relate this set to a
larger discourse frame. Countability and number are visually represented within the images.
The framework is meant to serve as both a conceptual tool for learning (Kozulin, 2003) and a semiotic
mediator (Valsiner, 2007) for individuals’ use of language. Teachers and learners may map disparate
article uses to the framework’s overarching schema and thus cohere rules and exceptions into one
seamless, meaningful, comprehensible system.
References
Fillmore, C. J. (2007). Frame semantics. In V. Evans, B. K. Bergen, & J. Zinken (Eds.), The cognitive
linguistics reader (pp. 238-262). London: Equinox.
Kozulin, A. (2003). Psychological tools and mediated learning. In A. Kozulin, B. Gindis, V. S. Ageyev,
& S. M. Miller (Eds.), Vygotsky’s educational theory in cultural context (pp. 15-38).
Cambridge: Cambridge University Press.
Langacker, R. (2008). Cognitive grammar: A basic introduction. Oxford: Oxford University Press.
Lantolf, J. P., & Thorne S. L. (2006). Sociocultural theory and the genesis of second language
development. Oxford: Oxford University Press.
Negueruela, E. (2009). A conceptual approach to promoting L2 grammatical development:
Implications for language program directors. In J. Watzinger-Tharp & S. Katz (Eds.), AAUSC
2008: Conceptions of L2 grammar: Theoretical approaches and their application in the L2
classroom (pp. 145-165). Florence, KY: Cengage Learning.
Pica, T. (1983). The article in American English: What the textbooks don’t tell us. In N. Wolfson & E.
Judd (Eds.), Sociolinguistics and language acquisition (pp. 222-233). Rowley, MA: Newbury
House.
Steen, G. J. (2007). Finding metaphor in grammar and usage. Amsterdam: John Benjamins.
Valsiner, J. (2007). Culture in minds and societies: Foundations of cultural psychology. London: Sage
Publications.
A Corpus-based Study of Chinese Non-basic Color Terms from the Perspective of Cognitive
Semantics
Yahong Xue
Jilin University, P. R. China
Color terms, as the products of human perception, categorization and encoding of color with natural
language, are words and expressions denoting the color attributes of things, including hue, brightness
and saturation and form a relatively independent yet dynamic lexical system of natural language,
which were viewed as best-case argument for linguistic relativity prior to Berlin and Kay’s universalist
tradition and remain one of the leading issues in the linguistic studies of the day. This paper makes a
corpus-based study from the perspective of cognitive semantics of Chinese non-basic color terms, the
subcategories of color terms, aiming to answer the following two research questions:
1. What are the representation forms of Chinese non-basic color terms?
2. What are the cognitive motivations involved in the naming of Chinese non-basic color terms?
The present study takes as its research methodology the combination of quantitative research and
qualitative research. The data used are taken from Corpus of Chinese Language(CCL). It is found
that the non-basic color terms, as subcategories of basic color terms, are formed in the cognitive
process of transcategorization, and there are four main representation forms of non-basic color terms
in Chinese: object color terms，proper noun color terms, color terms denoting degree, color terms
denoting mixed color. It is also found that the major cognitive motivations of the Chinese non-basic
color terms are metaphor, metonymy and conceptual integration.
The present study supports the new interpretation of Sapir-Whorf Hypothesis by Levinson(1997;
2003), namely, there is universality at the atomic level but differences at the molecular level in human
languages. The non-basic color terms directly reflect how a speech community construes color
experience as meaning and reveal the unique thinking mode of an ethnic group in getting to know
things as well as the close relation between linguistic categories, cognitive capacity and ethnic
traditions.
References
Berlin, B. & Kay, P. Basic Color Terms: Their Universality and Evolution[M]. Berkeley: University of
California Press，1969/1991.
Biggam, C. P. The Semantics of Colour: A Historical Approach[M]. Cambridge: Cambridge
University Press, 2012.
Kay, P. & McDaniel, C.K. The Linguistic Significance of the Meanings of Basic Color Terms,
Language, 1978.
Lakoff, G. & Johnson, M. Metaphors We Live By [M]. Chicago: University of Chicago Press, 1980.
Levinson, S. C. From outer to inner space: linguistic categories and non-linguistic thinking. In Nuyts,
J. & E. Pederson (Eds.) Language and Conceptualization [C]. Cambridge: Cambridge
University
Press, 1997: 13-45.
Levinson, S. C. Space in Language and Cognition[M]. England, Cambridge: Cambridge University
Press, 2003.
MacLaury, Robert E. Introducing vantage theory. Language Sciences, 2002: 24.5–6.493–536.
Corpus of Chinese Language(CCL) http://ccl.pku.edu.cn:8080/ccl_corpus/
Processing of Inflectional morphology in L1 and L2 Japanese
1
2
Peng Zhang , Hong Zhu
1,2
Japanese Department, Zhongnan University of Economics and Law, China
2
Tohoku University, Japan
The processing of Inflectional morphology was a highly-focused issue during the last two decades in
psycholinguistic. Previous studies in English demonstrated that the representation of English inflected
words either via a single route or a dual route processing by using a rule-based computation or a
whole-word recognition. (Marslen-Wilson, & Tyler, 1998; Pinker, 1999; Ullman, 2001) Differ from
English studies, there is little agreement on the mechanisms of inflectional morphology processing in
inflection-riched languages as a native language or a second language, and also on the exact roles of
morphological regularity, orthographical similarity and other morpho-syntax factors. (Gor, & cook,
2010; Clahsen, Felser, Sato, & Silva, 2010)
In this study, we used a cross-modal priming paradigm to explore the processing of inflected verbs
as well as the roles of morphological regularity, orthographical similarity in L1 and L2 Japanese, a
language with complex morphological transformation system. Materials included regular/irregular
inflected verbs, orthographically distinguished from similar to dissimilar by the criterion of Allen &
Badecker (2002). Each verb will match three prime conditions, the regular condition (kimerareru kimeru), the irregular condition (kimaru - kimeru), and the non-relative condition (sakkura - kimeru) as
a control prime to compare the facilitation between regular and irregular conditions.
The result indicated that native speakers of Japanese revealed greater facilitation for regular
inflections than irregular inflections, and were consistent with the dual-route system. In comparison,
L2 Japanese learners performed regular inflections faster than irregular inflections only in high
orthographical similarity conditions (kimaru - kimeru), but not in Low orthographical similarity
conditions (ochiru - otosu). We account for this result in terms of the low distinction of regular/irregular
inflective forms in Japanese. Unlike English, Japanese has a comparatively complex inflectional
system, inflections ranging from regular to irregular according to some potential rules of affix types.
(okutsu 1968; kageyama 2001) L2 Japanese Learners cannot grasp these potential rules as well as
natives do, so they tend to overuse the regular rules to the irregular forms. For this reason, the
difference of reaction times between regular and irregular forms was largely reduced, and
decomposed ways are in charge of the L2 inflection processing. Overall, the finding suggested that
inflective rules and orthographical similarity play an important role in processing of Japanese inflection
in L2 learners.
References
Allen, M. & W. Badecker. Inflectional Regularity: Probing the Nature of Lexical Representation in
a Cross-modalPriming Task [J]. Journal of Memory and Language, 2002, 46(4): 705-722
Clahsen, H., Felser, C., Neubauer, K., Sato, M., & Silva, R. Morphological Structure in Native and
Non-Native Language Processing [J]. Language Learning,2010, 60: 21-43.
Gor, K. & S. Cook. Nonnative processing of verbal morphology: In search of regularity [J].
Language Learning, 2010, 60(1): 88-126.
Marslen-Wilson, W.D. & L. K. Tyler. Rules, rep-resentations, and the English past Tense [J].
Trends in Cognitive Sciences, 1998, 11: 428-435.
Pinker, S. Words and rules: The ingredients of language [M]. NewYork: Basic Books, 1999.
Ullman, M. T. The neural basis of lexicon and grammar in first and second language: The
declarative / proceduralmodel [J]. Language and Cognition, 2001, 4(1): 105-122
Metaphorical Categorization and Its Dynamic Realizing Process
Rong Zhou
South China Normal University
The relation between metaphor and categorization has been an important issue in various
research fields, including philosophy, logics, linguistics, and cognitive science. The role of metaphor in
category extension is recognized as one of the basic tenets of cognitive linguistics and the
Embodiment Philosophy. Studies on the role of metaphor in categorization have significant theoretical
implications to our understanding of why metaphor is possible and how category can be fluid.
Previous cognitive linguistic studies were merely based on descriptive data on concrete language
forms, failing to systematically investigate the underlying mechanism. Hence, there still lacks a
comprehensive theoretical framework to account for the role of metaphor in category extension. The
present paper attempts to explore the process in which the metaphor topic and vehicle leap over the
conventional categorical boundaries and join a common superordinate category as metaphorical
categorization . Two experiments were conducted to answer two questions:
1, What is the characteristic of the time course in which the metaphor vehicle obtains an abstract
categorical referent in the course of metaphor processing?
2, What is the characteristic of the process in which the metaphor vehicle dynamically obtains an
abstract categorical referent in a vitro diachronical setting?
The first experiment employed priming metaphor similarity manipulation. A 2（metaphor prime
type: attributive metaphor prime, structural metaphor prime）×3（target type: concrete literal, abstract
categorical, unrelated）×3（target SOA：375ms,750ms,1500ms）mixed design was employed. 30
pairs of metaphors was taken from previous researches or generated by the author herself. Each pair
was designed to have the same vehicle yet different topics (e.g., “湖面-镜子”（lake surface- mirror）
and “历史-镜子” (history- mirror). It was intended that in each pair, one metaphor has attributive
similarity between the topic and vehicle (e.g., lake surface- mirror) while the other had relational
similarity (e.g., history- mirror). The participants were instructed to carefully read and figure out the
similarity conveyed by the topic and vehicle pair, and make a key press as quickly and accurately as
possible as to whether the target word was a meaningful Chinese word or not.
The second experiment adopted a diachronical paradigm. A 2（similarity of word pair: attributive
similar, structural similar）×2 （learning type: repeated topic, variant topic）×3（probe type: abstract
categorical, concrete literal , unrelated control）mixed design was used. Similarity of word pair and
learning type were manipulated between subjects while probe word type was varied within subjects.
The whole experiment consisted of two phases: a learning phase and a testing phase. The learning
phase was taken as the manipulation of the diachronical process of the categorization in the vehicle,
and the test phase in which a semantic priming paradigm was adopted was taken as the investigation
into the abstraction level of the vehicle.
Through the two experiments, we found that the induction of the abstract schema at the later
stage was due to the structure mapping at the earlier stage, and the time course of metaphorical
categorization had the “first-alignment-later-induction” characteristic; and the induction of the abstract
referent in the vehicle did not result from mere repeated memorization of the same topic-vehicle pair,
but from repeated activation of the same schema between the same vehicle and various topics. We
conclude that it is structure-mapping and schema induction that underlie the mechanism of
metaphorical categorization.